Original Link: https://www.anandtech.com/show/14292/shuttle-xpc-slim-dh370-minipc-review-a-compact-digital-signage-powerhouse
Shuttle XPC slim DH370 mini-PC Review: A Compact Digital Signage Powerhouse
by Ganesh T S on May 6, 2019 8:00 AM EST
Small form-factor (SFF) machines have predominantly used 5W - 45W TDP processors. The introduction of the mini-STX form-factor standardized a compact form factor for systems with processors having a TDP up to 65W. However, the mini-STX form-factor along with the associated cooling system is still too bulky for certain use-cases. Most systems built in that form-factor have considerable z-height to accommodate a reasonably standard heat-sink and a single fan. The resulting system volume is between 1.54L and 1.92L, depending on the chosen chassis.
Shuttle's XPC slim line aims to serve the market segment requiring standard desktop processors in a compact form factor. These computers are a fit for a variety of business and commercial use-cases. This review takes a look at the performance and features of a system built on the Shuttle XPC slim DH370 barebones platform.
Introduction and Platform Analysis
Shuttle's XPC slim line-up includes two different form-factors - the XH series uses a 17cm x 17cm motherboard in a 3.5L chassis, while the DH series uses a slightly smaller board in a 1.35L chassis. The DH370 we are looking at today belongs to the latter category. The product is sold 'barebones' similar to the mini-STX products in the market (and unlike the NUCs). The board is socketed, and the end user can opt to install any Coffee Lake-S CPU (up to 65W TDP) in the system. The RAM (DDR4 SO-DIMMs) and storage drive(s) (M.2 NVMe and/or 2.5" SATA drive) have to be added by the user. Shuttle supplies a dual-fan proprietary cooling solution along with the system. The choice of CPU makes it similar to a DIY build in many respects.
The barebones version of the DH370 is complemented by a rich set of optional accessories to better fit particular use-cases. These include the ability to replace one of the two COM ports with a VGA display output port (using the PVG-01), adding Wi-Fi capabilities (using the WLN-M WLAN card and antennae), ability to trigger power switching when the unit is installed in a hard-to-reach location (using the 2m long CXP01 cable for an external power switch), a rackmount kit allowing for the installation of two XPC slim units in a 2U slot (PRM01), and stands (PS02) for vertical orientation of the system.
Shuttle pre-configured our review sample with a Core i7-8700 processor, a Kingston A1000 PCIe 3.0 x2 NVMe SSD, and a HyperX Impact 2400 MHz 2x4GB SO-DIMM kit. An Intel Wireless-AC 9560 WLAN adapter was also pre-installed.
The specifications of our Shuttle XPC slim DH370 review configuration are summarized in the table below.
| Shuttle XPC slim DH370 Specifications | |
| Processor | Intel Core i7-8700 Coffee Lake-S, 6C/12T, 3.2 (4.6) GHz 12MB L2+L3, 65 W TDP |
| Memory | Kingston HyperX KHX2400C14S4/4G DDR4 SODIMM 14-14-14-35 @ 2400 MHz 2x4 GB |
| Graphics | Intel UHD Graphics 630 |
| Disk Drive(s) | Kingston A1000 SA1000M8240G (240 GB; M.2 2280 PCIe 3.0 x2; Toshiba 64L 3D TLC) |
| Networking | Intel Dual Band Wireless-AC 9560 (2x2 802.11ac - 1733 Mbps) 2x Intel i211 Gigabit Ethernet controller |
| Audio | 3.5mm Headphone / Microphone Jack (Realtek ALC662 audio codec) Capable of 5.1/7.1 digital output with HD audio bitstreaming (HDMI & DisplayPort) |
| Miscellaneous I/O Ports | 4x USB 3.1 Gen 2 Type-A 4x USB 3.0 Type-A 2x RS232 COM SD Card Reader (UHS-I) |
| Operating System | Retail unit is barebones, but we installed Windows 10 Enterprise x64 |
| Pricing | $330 (barebones) $759 (as configured, no OS) |
| Full Specifications | Shuttle XPC slim DH370 Barebones Specifications |
Similar to the other compact SFF systems, the DH370 is equipped with two DDR4 SO-DIMM slots (supporting DDR4-2400 / 2666 MHz memory kits). Since our review system was pre-configured, the package contents that we received and have pictured below might not exactly tally with the set of components received by retail customers.
The package includes the drivers on a CD (a USB key, even read-only, is much more preferable), a quick installation guide, screws to install the storage drives, thermal paste, a 90W (19V @ 4.74A) adapter, and a geo-specific power cord. The antennae for the WLAN component and the PS02 stand kit for vertical installation were also part of the kit. The gallery below shows aspects of the chassis design as well as the internals of the system.
The Shuttle XPC slim DH370 runs the standard AMI BIOS, and the features are quite basic. There is no fancy GUI or extra features such as the ones we are accustomed to in the BIOS of systems from Intel, ASRock, Asus, GIGABYTE, and the like. The screenshots in the gallery below takes us through the available BIOS options.
The DH370, as the name itself indicates, uses the Intel H370 chipset. This brings all the platform updates that the 8th Gen. Core platform supports - the primary ones being the plethora of USB 3.1 Gen 2 ports directly from the chipset, and the CNVi integrated Wi-Fi capabilities. The DH370 makes maximum use of the new features, opting to bring out all four possible USB 3.1 Gen 2 hosts as Type-A ports (two in the front panel, and two in the rear). The board also integrates a M.2 2230 slot for the Wi-Fi WLAN adapter. The H370 chipset has 20 PCIe 3.0 lanes, of which four are used for the DMI link to the processor. The compact nature of the XPC slim DH370 doesn't allow for complete usage of the available HSIO (high-speed I/O) lanes. The AIDA64 system report provides a breakdown of the usage of the PCIe lanes:
- PCIe 3.0 x1 port #5 In Use @ x1 (Intel I211 Gigabit Network Connection)
- PCIe 3.0 x1 port #6 In Use @ x1 (Intel I211 Gigabit Network Connection)
- PCIe 3.0 x1 port #8 In Use @ x1 (Realtek PCI-E Card Reader)
- PCIe 3.0 x4 port #9 In Use @ x4 (M.2 NVMe SSD Slot)
In the table below, we have an overview of the various systems that we are comparing the Shuttle XPC slim DH370 against. Note that they may not belong to the same market segment. The relevant configuration details of the machines are provided so that readers have an understanding of why some benchmark numbers are skewed for or against the Shuttle XPC slim DH370 when we come to those sections.
| Comparative PC Configurations | ||
| Aspect | Shuttle XPC slim DH370 | |
| CPU | Intel Core i7-8700 | Intel Core i7-8700 |
| GPU | Intel UHD Graphics 630 | Intel UHD Graphics 630 |
| RAM | Kingston HyperX KHX2400C14S4/4G DDR4 SODIMM 14-14-14-35 @ 2400 MHz 2x4 GB |
Kingston HyperX KHX2400C14S4/4G DDR4 SODIMM 14-14-14-35 @ 2400 MHz 2x4 GB |
| Storage | Kingston A1000 SA1000M8240G (240 GB; M.2 2280 PCIe 3.0 x2; Toshiba 64L 3D TLC) |
Kingston A1000 SA1000M8240G (240 GB; M.2 2280 PCIe 3.0 x2; Toshiba 64L 3D TLC) |
| Wi-Fi | Intel Dual Band Wireless-AC 9560 (2x2 802.11ac - 1733 Mbps) |
Intel Dual Band Wireless-AC 9560 (2x2 802.11ac - 1733 Mbps) |
| Price (in USD, when built) | $330 (barebones) $759 (as configured, No OS) |
$330 (barebones) $759 (as configured, No OS) |
The XPC slim DH370 stands out from the run-of-the-mill Coffee Lake desktops due to its ability to drive three separate 4Kp60 displays (1x HDMI 2.0a + 2x DisplayPort 1.2) using the integrated GPU in the installed CFL-S processor. The ability to support this feature without compromising on the form-factor is worthy of deeper analysis. We will be doing that further down in this review, in addition to the analysis of our usual benchmarks for SFF systems.
BAPCo SYSmark 2018
The Shuttle XPC slim DH370 was evaluated using our Fall 2018 test suite for small-form factor PCs. In the first section, we will be looking at SYSmark 2018.
BAPCo's SYSmark 2018 is an application-based benchmark that uses real-world applications to replay usage patterns of business users in the areas of productivity, creativity, and responsiveness. The 'Productivity Scenario' covers office-centric activities including word processing, spreadsheet usage, financial analysis, software development, application installation, file compression, and e-mail management. The 'Creativity Scenario' represents media-centric activities such as digital photo processing, AI and ML for face recognition in photos and videos for the purpose of content creation, etc. The 'Responsiveness Scenario' evaluates the ability of the system to react in a quick manner to user inputs in areas such as application and file launches, web browsing, and multi-tasking.
Scores are meant to be compared against a reference desktop (the SYSmark 2018 calibration system, a Dell Optiplex 5050 tower with a Core i3-7100 and 4GB of DDR4-2133 memory to go with a 128GB M.2 SATA III SSD). The calibration system scores 1000 in each of the scenarios. A score of, say, 2000, would imply that the system under test is twice as fast as the reference system.
SYSmark 2018 also adds energy measurement to the mix. A high score in the SYSmark benchmarks might be nice to have, but, potential customers also need to determine the balance between power consumption and the efficiency of the system. For example, in the average office scenario, it might not be worth purchasing a noisy and power-hungry PC just because it ends up with a 2000 score in the SYSmark 2014 SE benchmarks. In order to provide a balanced perspective, SYSmark 2018 also allows vendors and decision makers to track the energy consumption during each workload. In the graphs below, we find the total energy consumed by the PC under test for a single iteration of each SYSmark 2018 workload. For reference, the calibration system consumes 5.36 Wh for productivity, 7.71 Wh for creativity, 5.61 Wh for responsiveness, and 18.68 Wh overall.
The Core i7-8700 is a powerful hexa-core CPU, and it expectedly leads in a number of scenarios. The system lags in the responsiveness case, likely due to the usage of a PCIe 3.0 x2 SSD (the systems above it in that workload were all benchmarked with a PCIe 3.0 x4 NVMe SSD or an Optane storage drive). Overall, the reviewed Bean Canyon and Hades Canyon NUC configurations provide better performance numbers (albeit, at a much higher cost). On the energy consumption side, the system is somewhat on the back foot, thanks to the usage of a 65W TDP CPU as well as a powerful PCH (compared to the other reviewed systems). The usage of a PCIe 3.0 x2 SSD somewhat offsets this. However, the XPC slim DH370 configuration ends up being better than only the Skull Canyon and Hades Canyon NUCs in that metric.
UL Benchmarks - PCMark and 3DMark
This section deals with a selection of the UL Futuremark benchmarks - PCMark 10, PCMark 8, and 3DMark. While the first two evaluate the system as a whole, 3DMark focuses on the graphics capabilities.
PCMark 10
UL's PCMark 10 evaluates computing systems for various usage scenarios (generic / essential tasks such as web browsing and starting up applications, productivity tasks such as editing spreadsheets and documents, gaming, and digital content creation). We benchmarked select PCs with the PCMark 10 Extended profile and recorded the scores for various scenarios. These scores are heavily influenced by the CPU and GPU in the system, though the RAM and storage device also play a part. The power plan was set to Balanced for all the PCs while processing the PCMark 10 benchmark.
The performance of the DH370 is not as good as what one might expect from a system equipped with a Core i7-8700. In fact, we see it barely surpassing the DeskMini 310 equipped with a Core i3-8100. Two contributors to this overall issue are the use of a PCIe 3.0 x2 NVMe SSD (compared to the PCIe 3.0 x4 NVMe SSDs in almost all the systems above it in the benchmarks) and the usage of a DDR4-2400 kit (almost all other systems have the SODIMMs operating between 2666 MHz and 3000 MHz).
PCMark 8
We continue to present PCMark 8 benchmark results (as those have more comparison points) while our PCMark 10 scores database for systems grows in size. PCMark 8 provides various usage scenarios (home, creative and work) and offers ways to benchmark both baseline (CPU-only) as well as OpenCL accelerated (CPU + GPU) performance. We benchmarked select PCs for the OpenCL accelerated performance in all three usage scenarios. These scores are heavily influenced by the CPU and the memory sub-system.
The PCMark 8 numbers follow the PCMark 10 ordering, largely due to the same reasons discussed earlier.
3DMark
UL's 3DMark comes with a diverse set of graphics workloads that target different Direct3D feature levels. Correspondingly, the rendering resolutions are also different. We use 3DMark 2.4.4264 to get an idea of the graphics capabilities of the system. In this section, we take a look at the performance of the Shuttle XPC slim DH370 across the different 3DMark workloads.
3DMark Ice Storm
This workload has three levels of varying complexity - the vanilla Ice Storm, Ice Storm Unlimited, and Ice Storm Extreme. It is a cross-platform benchmark (which means that the scores can be compared across different tablets and smartphones as well). All three use DirectX 11 (feature level 9) / OpenGL ES 2.0. While the Extreme renders at 1920 x 1080, the other two render at 1280 x 720. The graphs below present the various Ice Storm worloads' numbers for different systems that we have evaluated.
| UL 3DMark - Ice Storm Workloads | |||
3DMark Cloud Gate
The Cloud Gate workload is meant for notebooks and typical home PCs, and uses DirectX 11 (feature level 10) to render frames at 1280 x 720. The graph below presents the overall score for the workload across all the systems that are being compared.
3DMark Sky Diver
The Sky Diver workload is meant for gaming notebooks and mid-range PCs, and uses DirectX 11 (feature level 11) to render frames at 1920 x 1080. The graph below presents the overall score for the workload across all the systems that are being compared.
3DMark Fire Strike Extreme
The Fire Strike benchmark has three workloads. The base version is meant for high-performance gaming PCs. Similar to Sky Diver, it uses DirectX 11 (feature level 11) to render frames at 1920 x 1080. The Ultra version targets 4K gaming system, and renders at 3840 x 2160. However, we only deal with the Extreme version in our benchmarking - It renders at 2560 x 1440, and targets multi-GPU systems and overclocked PCs. The graph below presents the overall score for the Fire Strike Extreme benchmark across all the systems that are being compared.
3DMark Time Spy
The Time Spy workload has two levels with different complexities. Both use DirectX 12 (feature level 11). However, the plain version targets high-performance gaming PCs with a 2560 x 1440 render resolution, while the Extreme version renders at 3840 x 2160 resolution. The graphs below present both numbers for all the systems that are being compared in this review.
| UL 3DMark - Time Spy Workloads | |||
3DMark Night Raid
The Night Raid workload is a DirectX 12 benchmark test. It is less demanding than Time Spy, and is optimized for integrated graphics. The graph below presents the overall score in this workload for different system configurations.
The numbers in the 3DMark workloads roughly tally with what can be expected from the Intel UHD Graphics 630 (also present in the Core i3-8100, which was used in the DeskMini 310 review). It is clear that performance in 3D / gaming workloads is not a strong point of the XPC slim DH370. Fortunately, Shuttle makes no mention of the gaming segment when discussing the target markets for the system.
Miscellaneous Performance Metrics - I
This section looks at some of the other commonly used benchmarks representative of the performance of specific real-world applications.
3D Rendering - CINEBENCH R15
We use CINEBENCH R15 for 3D rendering evaluation. The program provides three benchmark modes - OpenGL, single threaded and multi-threaded. Evaluation of different PC configurations in all three modes provided us the following results.
The excellent single-threaded performance of the Coffee Lake core, as well as the 6C/12T configuration come into play here in the single and multi-threaded scores. On the OpenGL side, the GPU is not particularly powerful, and the system comes in the bottom half of the graph.
x265 Benchmark
Next up, we have some video encoding benchmarks using x265 v2.8. The appropriate encoder executable is chosen based on the supported CPU features. In the first case, we encode 600 1080p YUV 4:2:0 frames into a 1080p30 HEVC Main-profile compatible video stream at 1 Mbps and record the average number of frames encoded per second.
Our second test case is 1200 4K YUV 4:2:0 frames getting encoded into a 4Kp60 HEVC Main10-profile video stream at 35 Mbps. The encoding FPS is recorded.
Again, the performance of the Core i7-8700 stands out, as its excellent single-threaded performance as well as the number of cores / threads help the XPC slim DH370 come out on top.
7-Zip
7-Zip is a very effective and efficient compression program, often beating out OpenCL accelerated commercial programs in benchmarks even while using just the CPU power. 7-Zip has a benchmarking program that provides tons of details regarding the underlying CPU's efficiency. In this subsection, we are interested in the compression and decompression rates when utilizing all the available threads for the LZMA algorithm.
In CPU-intensive benchmarks such as 7-zip (and x265 earlier), where the slower RAM doesn't have too much effect, the XPC slim DH370 with its Core i7-8700 comes out on top easily.
Cryptography Benchmarks
Cryptography has become an indispensable part of our interaction with computing systems. Almost all modern systems have some sort of hardware-acceleration for making cryptographic operations faster and more power efficient. In this sub-section, we look at two different real-world applications that may make use of this acceleration.
BitLocker is a Windows features that encrypts entire disk volumes. While drives that offer encryption capabilities are dealt with using that feature, most legacy systems and external drives have to use the host system implementation. Windows has no direct benchmark for BitLocker. However, we cooked up a BitLocker operation sequence to determine the adeptness of the system at handling BitLocker operations. We start off with a 2.5GB RAM drive in which a 2GB VHD (virtual hard disk) is created. This VHD is then mounted, and BitLocker is enabled on the volume. Once the BitLocker encryption process gets done, BitLocker is disabled. This triggers a decryption process. The times taken to complete the encryption and decryption are recorded. This process is repeated 25 times, and the average of the last 20 iterations is graphed below.
Despite the slower DRAM, the XPC slim DH370 emerges as the best performer for the BitLocker encryption and decryption workloads.
Creation of secure archives is best done through the use of AES-256 as the encryption method while password protecting ZIP files. We re-use the benchmark mode of 7-Zip to determine the AES256-CBC encryption and decryption rates using pure software as well as AES-NI. Note that the 7-Zip benchmark uses a 48KB buffer for this purpose.
The 7-zip encryption and decryption rates in the XPC slim DH370 are well ahead of the other systems being considered - a trend that we have seen across all the CPU-intensive benchmarks in this section.
Yet another cryptography application is secure network communication. OpenSSL can take advantage of the acceleration provided by the host system to make operations faster. It also has a benchmark mode that can use varying buffer sizes. We recorded the processing rate for a 8KB buffer using the hardware-accelerated AES256-CBC-HAC-SHA1 feature.
Here, we see the Bean Canyon NUC giving it close competition. The Ryzen system is well ahead of the Intel-based systems in this benchmark.
Miscellaneous Performance Metrics - II
Continuing our presentation of benchmarks of real-world workloads, this section discusses Agisoft's Photoscan and the Dolphin Emulator, followed by an evaluation of the storage aspect of the XPC slim DH370 configuration.
Agisoft Photoscan
Agisoft PhotoScan is a commercial program that converts 2D images into 3D point maps, meshes and textures. The program designers sent us a command line version in order to evaluate the efficiency of various systems that go under our review scanner. The command line version has two benchmark modes, one using the CPU and the other using both the CPU and GPU (via OpenCL). We present the results from our evaluation using the CPU mode only. The benchmark (v1.3) takes 84 photographs and does four stages of computation:
- Stage 1: Align Photographs (capable of OpenCL acceleration)
- Stage 2: Build Point Cloud (capable of OpenCL acceleration)
- Stage 3: Build Mesh
- Stage 4: Build Textures
We record the time taken for each stage. Since various elements of the software are single threaded, and others multithreaded, it is interesting to record the effects of CPU generations, speeds, number of cores, and DRAM parameters using this software.
The power of the Core i7-8700 again comes to the fore, as the XPC slim DH370 sweeps the workload across all stages comfortably despite the slower RAM.
Dolphin Emulator
Wrapping up our application benchmark numbers is the new Dolphin Emulator (v5) benchmark mode results. This is again a test of the CPU capabilities.
The XPC slim DH370 completes the benchmark in the least amount of time, followed closely by the Bean Canyon NUC. Overall, in CPU-intensive workloads, the compact form-factor of the XPC slim DH370 doesn't seem to affect the capabilities of the Core i7-8700.
Storage Performance
Storage performance is a major aspect which influences our experience with any computing system. On the storage side, one evaluation option would be the repetition of our strenuous SSD review tests on the drive(s) in the PC. Fortunately, to avoid that overkill, PCMark 8 has a storage bench where certain common workloads such as loading games and document processing are replayed on the target drive. Results are presented in two forms, one being a benchmark number and the other, a bandwidth figure. We ran the PCMark 8 storage bench on selected PCs and the results are presented below.
The XPC slim DH370 is handicapped by the use of a PCIe 3.0 x2 NVMe SSD (compared to the x4 SSDs in most of the other systems). Yet, the numbers are well behind the DeskMini A300, which also used a PCIe 3.0 x2 NVMe SSD (WD Blue NVMe). If the storage sub-system is important for the use-case, the end user can do much better than the Kingston A1000 that our review system was configured with. In fact, using the WD Blue NVMe SSD could provide better performance numbers at a lower cost. PCIe 3.0 x4 NVMe SSDs are also supported, if needed.
HTPC Credentials - Display Outputs Capabilities
The XPC slim DH370 comes with 3 display outputs, and their characteristics are summarized in the table below. From a HTPC use-case perspective, the entries of interest include the ability to support UHD (3840 x 2160) or higher resolutions, along with HDR and HDCP 2.2. The latter enables the display output to be used for viewing protected content such as 4K Netflix streams and play back UltraHD Blu-rays.
| Shuttle XPC slim DH370 Display Outputs | |||
| HDMI | DisplayPort 1 | DisplayPort 2 | |
| Version | 2.0a | 1.2 | |
| Max. Video Output | 3840x2160 @ 60Hz | 4096x2160 @ 60Hz | |
| HDCP | Yes (2.2) | ||
| HDR | Yes | ||
| HD Audio Bitstreaming | Yes | ||
The system allows all three display outputs to be simultaneously active at the maximum rated resolutions. To maintain uniformity, we tested all three outputs driving a 3840 x 2160 display at 59.94 Hz in 'extension' mode.
The system has no trouble running three 4Kp60 displays and doing casual browsing or animation displays on all three at the same time. We will be analyzing more interesting use-cases in a later section.
Supporting the display of high-resolution protected video content is a requirement for even a casual HTPC user. In addition, HTPC enthusiasts also want their systems to support refresh rates that either match or be an integral multiple of the frame rate of the video being displayed. Most displays / AVRs are able to transmit the supported refresh rates to the PC using the EDID metadata. In some cases, the desired refresh rate might be missing in the list of supported modes.
Custom Resolutions
Our evaluation of the Shuttle XPC slim DH370
Various display refresh rates ranging from 23.976 Hz to 59.94 Hz were tested. Of particular interest is the 23.976 Hz (23p) setting, which Intel used to have trouble with in the pre-Broadwell days.
The gallery below presents screenshots from the other refresh rates that were tested.
The system has no trouble maintaining a fairly accurate refresh rate throughout the duration of the video playback.
High Dynamic Range (HDR) Support
The ability of the system to support HDR output is brought out in the first line of the madVR OSD in the above pictures. The display / desktop was configured to be in HDR mode prior to the gathering of the above screenshots.
The ideal desktop experience needs a YCbCr 4:4:4 or RGB output in order to avoid chroma subsampling artifacts on text content in HDR mode. The Shuttle XPC slim DH370 has no issues with this aspect.
The CyberLink Ultra HD Blu-ray Advisor tool confirms that our setup (Shuttle XPC slim DH370 + Denon AVR-X3400H + TCL 55P607) supports HDCP 2.2 along with HDR. The PC also satisfies other requirements (such as SGX and appropriate Management Engine firmware) to enable playback of Ultra HD Blu-rays.
In fact, we were able to activate UHD Blu-ray playback on our standard HTPC setup, as well as the LG HDR 5K monitor driven using one of the DisplayPort outputs.
HTPC Credentials - YouTube and Netflix Streaming
Our HTPC testing with respect to YouTube had been restricted to playback of a 1080p music video using the native HTML5 player in Firefox. The move to 4K, and the need to evaluate HDR support have made us choose Mystery Box's Peru 8K HDR 60FPS video as our test sample moving forward. On PCs running Windows, it is recommended that HDR streaming videos be viewed using the Microsoft Edge browser after putting the desktop in HDR mode.
As expected, we get the VP9 Profile 2 4K HDR stream and Edge is able to play it back without any issues. Various metrics of interest such as GPU usage and at-wall power consumption were recorded for the first four minutes of the playback of the above video. The numbers are graphed below.
The GPU load is around 75% and the media engine load around 35%. At the wall, the system consumes around 35W in the steady state, with the GPU alone accounting for slightly less than 5W.
The Netflix 4K HDR capability works with native Windows Store app as well as the Microsoft Edge browser. We used the Windows Store app to evaluate the playback of Season 4 Episode 4 of the Netflix Test Patterns title. The OS screenshot facilities obviously can't capture the video being played back. However, the debug OSD (reachable by Ctrl-Alt-Shift-D) can be recorded.
The (hevc,hdr,prk) entry corresponding to the Video Track in the debug OSD, along with the A/V bitrate details (192 kbps / 16 Mbps) indicate that the HDR stream is indeed being played back. Similar to the YouTube streaming case, metrics such as GPU usage and at-wall power consumption were recorded for the first three minutes of the playback of the title. The numbers are graphed below.
In the steady state, the GPU and media engine loads are around 70%, with the hardware decoder being loaded at slightly south of 50%. The at-wall power is around 28W, withthe GPU alone contributing around 5W to that number.
Overall, the Shuttle XPC slim DH370 ticks all the boxes for OTT streaming capabilities. It is, however, not particularly power efficient while doing that.
HTPC Credentials - Local Media Playback and Video Processing
Evaluation of local media playback and video processing is done by playing back files encompassing a range of relevant codecs, containers, resolutions, and frame rates. A note of the efficiency is also made by tracking GPU usage and power consumption of the system at the wall. Users have their own preference for the playback software / decoder / renderer, and our aim is to have numbers representative of commonly encountered scenarios. Towards this, we played back the test streams using the following combinations:
- MPC-HC x64 1.8.5 + LAV Video Decoder (DXVA2 Native) + Enhanced Video Renderer - Custom Presenter (EVR-CP)
- MPC-HC x64 1.8.5 + LAV Video Decoder (D3D11) + madVR 0.92.17 (DXVA-Focused)
- MPC-HC x64 1.8.5 + LAV Video Decoder (D3D11) + madVR 0.92.17 (Lanczos-Focused)
- VLC 3.0.6
- Kodi 18.1
The thirteen test streams (each of 90s duration) were played back from the local disk with an interval of 30 seconds in-between. Various metrics including GPU usage and at-wall power consumption were recorded during the course of this playback. Prior to looking at the metrics, a quick summary of the decoding capabilities of the Intel UHD Graphics 630 is useful to have for context.
The GPU can provide hardware acceleration for all contemporary codecs.
All our playback tests were done with the desktop HDR setting turned on. It is possible for certain system configurations to have madVR automatically turn on/off the HDR capabilities prior to the playback of a HDR video, but, we didn't take advantage of that in our testing.
VLC and Kodi
VLC is the playback software of choice for the average PC user who doesn't need a ten-foot UI. Its install-and-play simplicity has made it extremely popular. Over the years, the software has gained the ability to take advantage of various hardware acceleration options. Kodi, on the other hand, has a ten-foot UI making it the perfect open-source software for dedicated HTPCs. Support for add-ons make it very extensible and capable of customization. We played back our test files using the default VLC and Kodi configurations, and recorded the following metrics.
| Video Playback Efficiency - VLC and Kodi | |||
VLC is able to play back all streams with hardware acceleration - GPU usage less than 80% and media engine load less than 40% - except for the VP9 Profile 2 video. At-wall power consumption ranges from around 20W for the 1080p24 H.264 stream to 45W for the 4Kp60 HEVC streams. In the software decode case, we have the at-wall number sooting up to 65W. Kodi 18.1, on the other hand, is uniformly good. All streams, including the VP9.2 one, are played back with hardware acceleration and no dropped frames.
The GPU and media engine usage do not cross 80% for any of the files. The at-wall power consumption is around 28W - 32W depending on the stream/
MPC-HC
MPC-HC offers an easy way to test out different combinations of decoders and renderers. The first configuration we evaluated is the default post-install scenario, with only the in-built LAV Video Decoder forced to DXVA2 Native mode. Two additional passes were done with different madVR configurations. In the first one (DXVA-focused), we configured madVR to make use of the DXVA-accelerated video processing capabilities as much as possible. In the second (Lanczos-focused), the image scaling algorithms were set to 'Lanczos 3-tap, with anti-ringing checked'. Chroma upscaling was configured to be 'BiCubic 75 with anti-ringing checked' in both cases. The metrics collected during the playback of the test files using the above three configurations are presented below.
| Video Playback Efficiency - MPC-HC with EVR-CP and madVR | |||
The experience with MPC-HC and the latest Intel drivers (v6709) couldn't be more starkly different from what we experienced with the Bean Canyon NUC. The VP9 Profile 2 stream apparently plays back with hardware acceleration, but, suffers from plenty of dropped frames. The GPU power consumption numbers also seem off-the-charts, compared to what we saw with Kodi. This translates to higher at-wall power consumption in the 50W range even for streams that are decoded with hardware acceleration. Given this state with the lean EVR-CP renderer, the less said about the madVR configurations, the better.
Overall, Kodi 18.1 turned out to be the perfect media player application for use in the Shuttle XPC slim DH370.
Driving Three 4K Displays - There is No Free Lunch
The unique feature of the Shuttle XPC slim DH370 is the ability of the system to drive three simultaneous 4Kp60 displays from the integrated GPU. This is particularly useful for digital signage purposes. However, if one were to purchase the DH370, expecting it to playback three different 4Kp60 video streams on three different 4K monitors, the person is in for some disappointment. In fact, I actually tried playing back three different 4K streams on the three monitors, and ended up with stuttering videos in all three.
In order to study this aspect further, I fired up DXVAChecker's benchmark section and benchmarked each video in our decode / rendering test set for playback at 3840x2160. The first iteration was done with one monitor connected, the second with two, and the final one with three 4Kp60 displays being driven by the iGPU.
Benchmarking the VP9 Profile 2 HDR Stream with DXVA Checker
The best-case playback frame rate obtained for each video in the three different display configurations is presented in the table below.
| Shuttle XPC slim DH370 DXVA Checker Playback (3840x2160) Benchmark Best-case FPS |
|||
| Single Monitor | Dual Monitor | Triple Monitor | |
| 480i60 MPEG2 | 118.2 | 91.9 | 83.3 |
| 576i50 H.264 | 124.1 | 103.4 | 98.4 |
| 720p60 H.264 | 101.5 | 100.4 | 78 |
| 1080i60 MPEG2 | 114.2 | 102.7 | 91.4 |
| 1080i60 H.264 | 104.7 | 92.6 | 82 |
| 1080i60 VC1 | 115.6 | 109.5 | 98.1 |
| 1080p60 H.264 | 108.1 | 86.7 | 82.6 |
| 1080p24 H.264 | 107.8 | 97.3 | 78.8 |
| 4Kp30 H.264 | 91.1 | 82.8 | 75.5 |
| 4Kp60 HEVC | 84.8 | 67.1 | 61.1 |
| 4Kp60 HEVC Main10 | 66.6 | 68.3 | 55.6 |
| 4Kp25 HEVC HDR | 55.5 | 52.9 | 49 |
| 4Kp60 VP9 Profile 2 HDR | 55.3 | 51.2 | 47.2 |
The above numbers indicate that playback with a single monitor is fine for all the codecs (ignoring the VP9.2 video, as DXVAChecker was not able to use the ff-vp9-d3d11va decoder for the benchmarking). However, once more monitors are added to the fray, the ability of the iGPU to decode and play back the video at the required frame rate comes into question.
To be clear, the iGPU does not seem to have the ability to decode three simultaneous 60fps streams for any of the tested codecs irrespective of the resolution - the best case appears to be 124 fps for the 576i H.264 stream in the single monitor mode. That said, decoding three simultaneous 24 fps streams for certain codecs or some legal combination based on the numbers in the above table seems to be possible.
Driving multiple displays from a discrete GPU is relatively simple. The GPU has its own RAM where the frame buffers can reside. The GTX 1650, with its 128-bit memory bus, has a 128GBps memory bandwidth number. Contrast that with the Coffee Lake desktop processors, which, in dual-channel mode, have less than 40GBps available for both the CPU and the GPU together. It is likely that driving three 4Kp60 displays can take up a significant chunk of the available bandwidth, resulting in the performance loss that we see above.
The takeaway is that the Shuttle XPC slim DH370 can drive three 4Kp60 displays simultaneously. However, users must be prepared for some performance loss in this process.
Miscellaneous Aspects
We have taken a detailed look at the performance of the Shuttle XPC slim DH370 and its suitability for different use-cases in its target markets. Prior to providing some concluding remarks, it is important to get an idea of the power consumption numbers for the system as well as the efficiency of its cooling solution.
Power Consumption
The power consumption at the wall was measured with a 4K display being driven through the HDMI port. In the graphs below, we compare the idle and load power of the Shuttle XPC slim DH370 with other low power PCs evaluated before. For load power consumption, we ran the AIDA64 System Stability Test with various stress components, as well as our custom Prime95 / Furmark loading scripts, and noted the maximum sustained power consumption at the wall.
The system's 10.44W idling power is praiseworthy, given the choice of components. Stressing the 65W TDP processor also makes the system sustain more than 97W at the wall. This is in contrast to the 69W stress number for the 65W TDP Core i3-8100 in the DeskMini 310. The design of the Shuttle XPC slim allows for taking full advantage of the capabilities of the installed processor.
Thermal Performance
Our thermal stress routine starts with the system at idle, followed by four stages of different system loading profiles using the AIDA64 System Stability Test (each of 30 minutes duration). In the first stage, we stress the CPU, caches and RAM. In the second stage, we add the GPU to the above list. In the third stage, we stress the GPU standalone. In the final stage, we stress all the system components (including the disks). Beyond this, we leave the unit idle in order to determine how quickly the various temperatures in the system can come back to normal idling range. The various clocks, temperatures and power consumption numbers for the system during the above routine are presented in the graphs below.
| Shuttle XPC slim DH370 System Loading with the AIDA64 System Stability Test | |||
Despite being rated for operation at 3.2 GHz, the cores spend a majority of their time around the 3 GHz mark in the course of the AIDA stress test. The GPU operates at up to 1.2 GHz. The temperatures always stay south of 85C. The power numbers are more interesting. When all components are getting stressed, the CPU package power is constant around 48W. The interesting aspect to note here is that there is not much variation in the package power for each workload component.
We also run a custom stress test involving Prime95 and Furmark. Starting with Prime95 alone, we add Furmark to the mix after 30 minutes. After another 30 minutes of simultaneous CPU and GPU loading, we terminate the Prime95 process alone and let the GPU run at full throttle. The metrics graphed for the AIDA64 system stability test are also graphed here.
| Shuttle XPC slim DH370 System Loading with Prime95 and Furmark | |||
Our custom test is much more stressful. We see the cores starting out at 3.2 GHz, but, Prime95 makes them reach 90C around 20 minutes into the test. Once the temperature goes above, we see the cores clock down to around 3 GHz. Despite the lower clocks, the package power actually rises from around 50W to 52W. The addition of Furmark to the mix brings down the package power to 45W, and running only Furmark results in a package power dissipation of around 40W. Obviously, the latter two cases are much easier to handle for the cooling solution and we see the temperatures dropping rapidly.
Concluding Remarks
The Shuttle XPC slim DH370 is a compact computing powerhouse. All our benchmark numbers point to that. Shuttle's choice of RAM and SSD for our review configuration actually hold back the numbers a bit. Upgrading the RAM kit (to, say, DDR4-2666) and the SSD (to, say, a PCIe 3.0 x4 NVMe one) would only serve to make the already chart-leading performance better. The choice of external high-speed I/O - in particular, making all four possible USB 3.1 Gen 2 ports come out, is laudable. Dual Intel NICs ought to make the DH370 a good fit for certain networking use-cases.
Shuttle targets various market segments, as shown in their product brief. The DH370 is a very good fit for all of them, but, with caveats. The system can indeed drive three different 4Kp60 displays with HDR simultaneously. However, end-users should note that the usage of this mode may result in performance loss for some of the system workloads. For in-store and home media playback and entertainment, the XPC slim DH370 ticks all the right boxes, but, the fan noise is a definite factor. It is possible that a T-series processor might not stress the cooling system as much. However, the configuration we evaluated was simply too noisy for use as a HTPC. The drawbacks we encountered are not much of a factor for the other targeted applications.
The Shuttle XPC slim DH370 deserves a lot of praise for incorporating a balanced set of features available in the Intel H370 platform into a compact system. The thermal solution (pictured above) does a lot of the heavy lifting to allow the installed processor to perform to its potential. However, its noise profile is also the one trade-off that consumers have to make for the compact size of the system. The $330 barebones price for the system is also reasonable considering the standalone prices for a good H370 motherboard, chassis, and PSU.
Shuttle does have scope for improving the DH370 further - for starters, we would have liked a couple of the USB 3.1 Gen 2 Type-A ports to be Type-C. A Thunderbolt 3 port would have also been nice to have, given that spare PCIe lanes from the PCH as well as the CPU are available. That said, Shuttle's unique features (such as the triple 4K display output) are quite difficult to achieve in a DIY build in this form factor. Based on the results of our evaluation, we recommend the Shuttle XPC slim DH370 for purchase after careful analysis of the expected use-cases.
Facebook
Twitter
RSS
